*/
#define ASSEMBLY 1
-#include <linux/config.h>
#include <linux/linkage.h>
#include <asm/segment.h>
#include <asm/smp.h>
#include <asm/thread_info.h>
#include <asm/hw_irq.h>
#include <asm/page.h>
+#include <asm/irqflags.h>
.code64
#ifndef CONFIG_PREEMPT
#define retint_kernel retint_restore_args
#endif
-
+
+
+.macro TRACE_IRQS_IRETQ offset=ARGOFFSET
+#ifdef CONFIG_TRACE_IRQFLAGS
+ bt $9,EFLAGS-\offset(%rsp) /* interrupts off? */
+ jnc 1f
+ TRACE_IRQS_ON
+1:
+#endif
+.endm
+
/*
* C code is not supposed to know about undefined top of stack. Every time
* a C function with an pt_regs argument is called from the SYSCALL based
GET_THREAD_INFO(%rcx)
jmp rff_action
CFI_ENDPROC
+END(ret_from_fork)
/*
* System call entry. Upto 6 arguments in registers are supported.
ENTRY(system_call)
CFI_STARTPROC simple
- CFI_DEF_CFA rsp,0
+ CFI_DEF_CFA rsp,PDA_STACKOFFSET
CFI_REGISTER rip,rcx
/*CFI_REGISTER rflags,r11*/
swapgs
movq %rsp,%gs:pda_oldrsp
movq %gs:pda_kernelstack,%rsp
+ /*
+ * No need to follow this irqs off/on section - it's straight
+ * and short:
+ */
sti
SAVE_ARGS 8,1
movq %rax,ORIG_RAX-ARGOFFSET(%rsp)
sysret_check:
GET_THREAD_INFO(%rcx)
cli
+ TRACE_IRQS_OFF
movl threadinfo_flags(%rcx),%edx
andl %edi,%edx
CFI_REMEMBER_STATE
jnz sysret_careful
+ /*
+ * sysretq will re-enable interrupts:
+ */
+ TRACE_IRQS_ON
movq RIP-ARGOFFSET(%rsp),%rcx
CFI_REGISTER rip,rcx
RESTORE_ARGS 0,-ARG_SKIP,1
CFI_RESTORE_STATE
bt $TIF_NEED_RESCHED,%edx
jnc sysret_signal
+ TRACE_IRQS_ON
sti
pushq %rdi
CFI_ADJUST_CFA_OFFSET 8
/* Handle a signal */
sysret_signal:
+ TRACE_IRQS_ON
sti
testl $(_TIF_SIGPENDING|_TIF_NOTIFY_RESUME|_TIF_SINGLESTEP),%edx
jz 1f
/* Use IRET because user could have changed frame. This
works because ptregscall_common has called FIXUP_TOP_OF_STACK. */
cli
+ TRACE_IRQS_OFF
jmp int_with_check
badsys:
/* Use IRET because user could have changed frame */
jmp int_ret_from_sys_call
CFI_ENDPROC
+END(system_call)
/*
* Syscall return path ending with IRET.
CFI_REL_OFFSET r10,R10-ARGOFFSET
CFI_REL_OFFSET r11,R11-ARGOFFSET
cli
+ TRACE_IRQS_OFF
testl $3,CS-ARGOFFSET(%rsp)
je retint_restore_args
movl $_TIF_ALLWORK_MASK,%edi
int_careful:
bt $TIF_NEED_RESCHED,%edx
jnc int_very_careful
+ TRACE_IRQS_ON
sti
pushq %rdi
CFI_ADJUST_CFA_OFFSET 8
popq %rdi
CFI_ADJUST_CFA_OFFSET -8
cli
+ TRACE_IRQS_OFF
jmp int_with_check
/* handle signals and tracing -- both require a full stack frame */
int_very_careful:
+ TRACE_IRQS_ON
sti
SAVE_REST
/* Check for syscall exit trace */
CFI_ADJUST_CFA_OFFSET -8
andl $~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edi
cli
+ TRACE_IRQS_OFF
jmp int_restore_rest
int_signal:
int_restore_rest:
RESTORE_REST
cli
+ TRACE_IRQS_OFF
jmp int_with_check
CFI_ENDPROC
+END(int_ret_from_sys_call)
/*
* Certain special system calls that need to save a complete full stack frame.
leaq \func(%rip),%rax
leaq -ARGOFFSET+8(%rsp),\arg /* 8 for return address */
jmp ptregscall_common
+END(\label)
.endm
CFI_STARTPROC
CFI_REL_OFFSET rip, 0
ret
CFI_ENDPROC
+END(ptregscall_common)
ENTRY(stub_execve)
CFI_STARTPROC
RESTORE_REST
jmp int_ret_from_sys_call
CFI_ENDPROC
+END(stub_execve)
/*
* sigreturn is special because it needs to restore all registers on return.
RESTORE_REST
jmp int_ret_from_sys_call
CFI_ENDPROC
+END(stub_rt_sigreturn)
/*
* initial frame state for interrupts and exceptions
/* 0(%rsp): interrupt number */
.macro interrupt func
cld
-#ifdef CONFIG_DEBUG_INFO
- SAVE_ALL
- movq %rsp,%rdi
- /*
- * Setup a stack frame pointer. This allows gdb to trace
- * back to the original stack.
- */
- movq %rsp,%rbp
- CFI_DEF_CFA_REGISTER rbp
-#else
SAVE_ARGS
leaq -ARGOFFSET(%rsp),%rdi # arg1 for handler
-#endif
+ pushq %rbp
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rbp, 0
+ movq %rsp,%rbp
+ CFI_DEF_CFA_REGISTER rbp
testl $3,CS(%rdi)
je 1f
swapgs
1: incl %gs:pda_irqcount # RED-PEN should check preempt count
- movq %gs:pda_irqstackptr,%rax
- cmoveq %rax,%rsp /*todo This needs CFI annotation! */
- pushq %rdi # save old stack
-#ifndef CONFIG_DEBUG_INFO
- CFI_ADJUST_CFA_OFFSET 8
-#endif
+ cmoveq %gs:pda_irqstackptr,%rsp
+ push %rbp # backlink for old unwinder
+ /*
+ * We entered an interrupt context - irqs are off:
+ */
+ TRACE_IRQS_OFF
call \func
.endm
interrupt do_IRQ
/* 0(%rsp): oldrsp-ARGOFFSET */
ret_from_intr:
- popq %rdi
-#ifndef CONFIG_DEBUG_INFO
- CFI_ADJUST_CFA_OFFSET -8
-#endif
cli
+ TRACE_IRQS_OFF
decl %gs:pda_irqcount
-#ifdef CONFIG_DEBUG_INFO
- movq RBP(%rdi),%rbp
+ leaveq
CFI_DEF_CFA_REGISTER rsp
-#endif
- leaq ARGOFFSET(%rdi),%rsp /*todo This needs CFI annotation! */
+ CFI_ADJUST_CFA_OFFSET -8
exit_intr:
GET_THREAD_INFO(%rcx)
testl $3,CS-ARGOFFSET(%rsp)
CFI_REMEMBER_STATE
jnz retint_careful
retint_swapgs:
+ /*
+ * The iretq could re-enable interrupts:
+ */
+ cli
+ TRACE_IRQS_IRETQ
swapgs
+ jmp restore_args
+
retint_restore_args:
cli
+ /*
+ * The iretq could re-enable interrupts:
+ */
+ TRACE_IRQS_IRETQ
+restore_args:
RESTORE_ARGS 0,8,0
iret_label:
iretq
/* running with kernel gs */
bad_iret:
movq $11,%rdi /* SIGSEGV */
+ TRACE_IRQS_ON
sti
jmp do_exit
.previous
CFI_RESTORE_STATE
bt $TIF_NEED_RESCHED,%edx
jnc retint_signal
+ TRACE_IRQS_ON
sti
pushq %rdi
CFI_ADJUST_CFA_OFFSET 8
CFI_ADJUST_CFA_OFFSET -8
GET_THREAD_INFO(%rcx)
cli
+ TRACE_IRQS_OFF
jmp retint_check
retint_signal:
testl $(_TIF_SIGPENDING|_TIF_NOTIFY_RESUME|_TIF_SINGLESTEP),%edx
jz retint_swapgs
+ TRACE_IRQS_ON
sti
SAVE_REST
movq $-1,ORIG_RAX(%rsp)
call do_notify_resume
RESTORE_REST
cli
+ TRACE_IRQS_OFF
movl $_TIF_NEED_RESCHED,%edi
GET_THREAD_INFO(%rcx)
jmp retint_check
call preempt_schedule_irq
jmp exit_intr
#endif
+
CFI_ENDPROC
+END(common_interrupt)
/*
* APIC interrupts.
ENTRY(thermal_interrupt)
apicinterrupt THERMAL_APIC_VECTOR,smp_thermal_interrupt
+END(thermal_interrupt)
ENTRY(threshold_interrupt)
apicinterrupt THRESHOLD_APIC_VECTOR,mce_threshold_interrupt
+END(threshold_interrupt)
#ifdef CONFIG_SMP
ENTRY(reschedule_interrupt)
apicinterrupt RESCHEDULE_VECTOR,smp_reschedule_interrupt
+END(reschedule_interrupt)
.macro INVALIDATE_ENTRY num
ENTRY(invalidate_interrupt\num)
apicinterrupt INVALIDATE_TLB_VECTOR_START+\num,smp_invalidate_interrupt
+END(invalidate_interrupt\num)
.endm
INVALIDATE_ENTRY 0
ENTRY(call_function_interrupt)
apicinterrupt CALL_FUNCTION_VECTOR,smp_call_function_interrupt
+END(call_function_interrupt)
#endif
#ifdef CONFIG_X86_LOCAL_APIC
ENTRY(apic_timer_interrupt)
apicinterrupt LOCAL_TIMER_VECTOR,smp_apic_timer_interrupt
+END(apic_timer_interrupt)
ENTRY(error_interrupt)
apicinterrupt ERROR_APIC_VECTOR,smp_error_interrupt
+END(error_interrupt)
ENTRY(spurious_interrupt)
apicinterrupt SPURIOUS_APIC_VECTOR,smp_spurious_interrupt
+END(spurious_interrupt)
#endif
/*
/* error code is on the stack already */
/* handle NMI like exceptions that can happen everywhere */
- .macro paranoidentry sym, ist=0
+ .macro paranoidentry sym, ist=0, irqtrace=1
SAVE_ALL
cld
movl $1,%ebx
addq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp)
.endif
cli
+ .if \irqtrace
+ TRACE_IRQS_OFF
+ .endif
.endm
-
+
+ /*
+ * "Paranoid" exit path from exception stack.
+ * Paranoid because this is used by NMIs and cannot take
+ * any kernel state for granted.
+ * We don't do kernel preemption checks here, because only
+ * NMI should be common and it does not enable IRQs and
+ * cannot get reschedule ticks.
+ *
+ * "trace" is 0 for the NMI handler only, because irq-tracing
+ * is fundamentally NMI-unsafe. (we cannot change the soft and
+ * hard flags at once, atomically)
+ */
+ .macro paranoidexit trace=1
+ /* ebx: no swapgs flag */
+paranoid_exit\trace:
+ testl %ebx,%ebx /* swapgs needed? */
+ jnz paranoid_restore\trace
+ testl $3,CS(%rsp)
+ jnz paranoid_userspace\trace
+paranoid_swapgs\trace:
+ TRACE_IRQS_IRETQ 0
+ swapgs
+paranoid_restore\trace:
+ RESTORE_ALL 8
+ iretq
+paranoid_userspace\trace:
+ GET_THREAD_INFO(%rcx)
+ movl threadinfo_flags(%rcx),%ebx
+ andl $_TIF_WORK_MASK,%ebx
+ jz paranoid_swapgs\trace
+ movq %rsp,%rdi /* &pt_regs */
+ call sync_regs
+ movq %rax,%rsp /* switch stack for scheduling */
+ testl $_TIF_NEED_RESCHED,%ebx
+ jnz paranoid_schedule\trace
+ movl %ebx,%edx /* arg3: thread flags */
+ .if \trace
+ TRACE_IRQS_ON
+ .endif
+ sti
+ xorl %esi,%esi /* arg2: oldset */
+ movq %rsp,%rdi /* arg1: &pt_regs */
+ call do_notify_resume
+ cli
+ .if \trace
+ TRACE_IRQS_OFF
+ .endif
+ jmp paranoid_userspace\trace
+paranoid_schedule\trace:
+ .if \trace
+ TRACE_IRQS_ON
+ .endif
+ sti
+ call schedule
+ cli
+ .if \trace
+ TRACE_IRQS_OFF
+ .endif
+ jmp paranoid_userspace\trace
+ CFI_ENDPROC
+ .endm
+
/*
* Exception entry point. This expects an error code/orig_rax on the stack
* and the exception handler in %rax.
movl %ebx,%eax
RESTORE_REST
cli
+ TRACE_IRQS_OFF
GET_THREAD_INFO(%rcx)
testl %eax,%eax
jne retint_kernel
movl $_TIF_WORK_MASK,%edi
andl %edi,%edx
jnz retint_careful
+ /*
+ * The iret might restore flags:
+ */
+ TRACE_IRQS_IRETQ
swapgs
RESTORE_ARGS 0,8,0
jmp iret_label
cmpq $gs_change,RIP(%rsp)
je error_swapgs
jmp error_sti
+END(error_entry)
/* Reload gs selector with exception handling */
/* edi: new selector */
CFI_ADJUST_CFA_OFFSET -8
ret
CFI_ENDPROC
+ENDPROC(load_gs_index)
.section __ex_table,"a"
.align 8
call do_fork
movq %rax,RAX(%rsp)
xorl %edi,%edi
+ test %rax,%rax
+ jnz 1f
+ /* terminate stack in child */
+ movq %rdi,RIP(%rsp)
+1:
/*
* It isn't worth to check for reschedule here,
UNFAKE_STACK_FRAME
ret
CFI_ENDPROC
-
+ENDPROC(kernel_thread)
child_rip:
+ pushq $0 # fake return address
+ CFI_STARTPROC
/*
* Here we are in the child and the registers are set as they were
* at kernel_thread() invocation in the parent.
# exit
xorl %edi, %edi
call do_exit
+ CFI_ENDPROC
+ENDPROC(child_rip)
/*
* execve(). This function needs to use IRET, not SYSRET, to set up all state properly.
UNFAKE_STACK_FRAME
ret
CFI_ENDPROC
+ENDPROC(execve)
KPROBE_ENTRY(page_fault)
errorentry do_page_fault
+END(page_fault)
.previous .text
ENTRY(coprocessor_error)
zeroentry do_coprocessor_error
+END(coprocessor_error)
ENTRY(simd_coprocessor_error)
zeroentry do_simd_coprocessor_error
+END(simd_coprocessor_error)
ENTRY(device_not_available)
zeroentry math_state_restore
+END(device_not_available)
/* runs on exception stack */
KPROBE_ENTRY(debug)
pushq $0
CFI_ADJUST_CFA_OFFSET 8
paranoidentry do_debug, DEBUG_STACK
- jmp paranoid_exit
- CFI_ENDPROC
+ paranoidexit
+END(debug)
.previous .text
/* runs on exception stack */
INTR_FRAME
pushq $-1
CFI_ADJUST_CFA_OFFSET 8
- paranoidentry do_nmi
- /*
- * "Paranoid" exit path from exception stack.
- * Paranoid because this is used by NMIs and cannot take
- * any kernel state for granted.
- * We don't do kernel preemption checks here, because only
- * NMI should be common and it does not enable IRQs and
- * cannot get reschedule ticks.
- */
- /* ebx: no swapgs flag */
-paranoid_exit:
- testl %ebx,%ebx /* swapgs needed? */
- jnz paranoid_restore
- testl $3,CS(%rsp)
- jnz paranoid_userspace
-paranoid_swapgs:
- swapgs
-paranoid_restore:
- RESTORE_ALL 8
- iretq
-paranoid_userspace:
- GET_THREAD_INFO(%rcx)
- movl threadinfo_flags(%rcx),%ebx
- andl $_TIF_WORK_MASK,%ebx
- jz paranoid_swapgs
- movq %rsp,%rdi /* &pt_regs */
- call sync_regs
- movq %rax,%rsp /* switch stack for scheduling */
- testl $_TIF_NEED_RESCHED,%ebx
- jnz paranoid_schedule
- movl %ebx,%edx /* arg3: thread flags */
- sti
- xorl %esi,%esi /* arg2: oldset */
- movq %rsp,%rdi /* arg1: &pt_regs */
- call do_notify_resume
- cli
- jmp paranoid_userspace
-paranoid_schedule:
- sti
- call schedule
- cli
- jmp paranoid_userspace
- CFI_ENDPROC
+ paranoidentry do_nmi, 0, 0
+#ifdef CONFIG_TRACE_IRQFLAGS
+ paranoidexit 0
+#else
+ jmp paranoid_exit1
+ CFI_ENDPROC
+#endif
+END(nmi)
.previous .text
KPROBE_ENTRY(int3)
pushq $0
CFI_ADJUST_CFA_OFFSET 8
paranoidentry do_int3, DEBUG_STACK
- jmp paranoid_exit
+ jmp paranoid_exit1
CFI_ENDPROC
+END(int3)
.previous .text
ENTRY(overflow)
zeroentry do_overflow
+END(overflow)
ENTRY(bounds)
zeroentry do_bounds
+END(bounds)
ENTRY(invalid_op)
zeroentry do_invalid_op
+END(invalid_op)
ENTRY(coprocessor_segment_overrun)
zeroentry do_coprocessor_segment_overrun
+END(coprocessor_segment_overrun)
ENTRY(reserved)
zeroentry do_reserved
+END(reserved)
/* runs on exception stack */
ENTRY(double_fault)
XCPT_FRAME
paranoidentry do_double_fault
- jmp paranoid_exit
+ jmp paranoid_exit1
CFI_ENDPROC
+END(double_fault)
ENTRY(invalid_TSS)
errorentry do_invalid_TSS
+END(invalid_TSS)
ENTRY(segment_not_present)
errorentry do_segment_not_present
+END(segment_not_present)
/* runs on exception stack */
ENTRY(stack_segment)
XCPT_FRAME
paranoidentry do_stack_segment
- jmp paranoid_exit
+ jmp paranoid_exit1
CFI_ENDPROC
+END(stack_segment)
KPROBE_ENTRY(general_protection)
errorentry do_general_protection
+END(general_protection)
.previous .text
ENTRY(alignment_check)
errorentry do_alignment_check
+END(alignment_check)
ENTRY(divide_error)
zeroentry do_divide_error
+END(divide_error)
ENTRY(spurious_interrupt_bug)
zeroentry do_spurious_interrupt_bug
+END(spurious_interrupt_bug)
#ifdef CONFIG_X86_MCE
/* runs on exception stack */
pushq $0
CFI_ADJUST_CFA_OFFSET 8
paranoidentry do_machine_check
- jmp paranoid_exit
+ jmp paranoid_exit1
CFI_ENDPROC
+END(machine_check)
#endif
+/* Call softirq on interrupt stack. Interrupts are off. */
ENTRY(call_softirq)
CFI_STARTPROC
- movq %gs:pda_irqstackptr,%rax
- movq %rsp,%rdx
- CFI_DEF_CFA_REGISTER rdx
+ push %rbp
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rbp,0
+ mov %rsp,%rbp
+ CFI_DEF_CFA_REGISTER rbp
incl %gs:pda_irqcount
- cmove %rax,%rsp
- pushq %rdx
- /*todo CFI_DEF_CFA_EXPRESSION ...*/
+ cmove %gs:pda_irqstackptr,%rsp
+ push %rbp # backlink for old unwinder
call __do_softirq
- popq %rsp
+ leaveq
CFI_DEF_CFA_REGISTER rsp
+ CFI_ADJUST_CFA_OFFSET -8
decl %gs:pda_irqcount
ret
CFI_ENDPROC
+ENDPROC(call_softirq)
+
+#ifdef CONFIG_STACK_UNWIND
+ENTRY(arch_unwind_init_running)
+ CFI_STARTPROC
+ movq %r15, R15(%rdi)
+ movq %r14, R14(%rdi)
+ xchgq %rsi, %rdx
+ movq %r13, R13(%rdi)
+ movq %r12, R12(%rdi)
+ xorl %eax, %eax
+ movq %rbp, RBP(%rdi)
+ movq %rbx, RBX(%rdi)
+ movq (%rsp), %rcx
+ movq %rax, R11(%rdi)
+ movq %rax, R10(%rdi)
+ movq %rax, R9(%rdi)
+ movq %rax, R8(%rdi)
+ movq %rax, RAX(%rdi)
+ movq %rax, RCX(%rdi)
+ movq %rax, RDX(%rdi)
+ movq %rax, RSI(%rdi)
+ movq %rax, RDI(%rdi)
+ movq %rax, ORIG_RAX(%rdi)
+ movq %rcx, RIP(%rdi)
+ leaq 8(%rsp), %rcx
+ movq $__KERNEL_CS, CS(%rdi)
+ movq %rax, EFLAGS(%rdi)
+ movq %rcx, RSP(%rdi)
+ movq $__KERNEL_DS, SS(%rdi)
+ jmpq *%rdx
+ CFI_ENDPROC
+ENDPROC(arch_unwind_init_running)
+#endif
git://git.onelab.eu / linux-2.6.git / blobdiff